MapReduce vs. pipelining counting triangles
Document typeConference report
Rights accessOpen Access
All rights reserved. This work is protected by the corresponding intellectual and industrial property rights. Without prejudice to any existing legal exemptions, reproduction, distribution, public communication or transformation of this work are prohibited without permission of the copyright holder
ProjectMODELOS Y METODOS COMPUTACIONALES PARA DATOS MASIVOS ESTRUCTURADOS (MINECO-TIN2013-46181-C2-1-R)
In this paper we follow an alternative approach named pipeline, to implement a parallel implementation of the well-known problem of counting triangles in a graph. This problem is especially interesting either when the input graph does not fit in memory or is dynamically generated. To be concrete, we implement a dynamic pipeline of processes and an ad-hoc MapReduce version using the language Go. We explote the ability of Go language to deal with channels and spawned processes. An empirical evaluation is conducted on graphs of different size and density. Observed results suggest that pipeline allows for the implementation of an efficient solution of the problem of counting triangles in a graph, particularly, in dense and large graphs, drastically reducing the execution time with respect to the MapReduce implementation.
CitationEdelmira Pasarella, Maria-Esther Vidal, Cristina Zoltan. MapReduce vs. pipelining counting triangles. A: Alberto Mendelzon Workshop on Foundations of Data Management. "AMW2016: proceedings of the 10th Alberto Mendelzon International Workshop on Foundations of Data Management: Panama City, Panama, May 8-10, 2016". Panama City: CEUR-WS.org, 2016, p. 1-5.